Background: Conventional transcranial magnetic stimulation (TMS) neuronavigation systems provide precise guidance but depend on costly optical tracking equipment. In this study, we developed and evaluated a smartphone-based neuronavigation system that uses RGB images alone, without depth sensing, for motor evoked potential (MEP) mapping. Methods: Fifteen healthy participants underwent TMS mapping of the first dorsal interosseous muscle. A smartphone RGB camera and AprilTag-based tracking were used to estimate head and coil poses. Smartphone-derived coordinates were registered to three-dimensional head scans and transformed into Montreal Neurological Institute space using affine transformations. MEP amplitude maps and centers of mass (CoMs) were then calculated. A mannequin simulation and reanalysis of feasibility-study data were conducted to determine a practical control-point configuration. Results: Registration combining anatomical landmarks with target control points yielded MEP CoM coordinates comparable to those obtained with conventional neuronavigation, although localization errors were slightly greater. Most estimated CoMs were located in the left precentral gyrus. Additional analyses demonstrated that four spatially distributed control points reduced coordinate-estimation errors relative to anatomical landmarks alone while preserving a simple registration procedure. Conclusions: A smartphone-based neuronavigation system using RGB images alone can produce physiologically plausible MEP maps without depth sensors or multi-camera systems. Despite the need for greater localization accuracy, improved lighting robustness, and magnetic resonance imaging integration, this approach could offer an accessible and cost-effective platform for research and clinical TMS.